Publications (5)
Adam-HNAG: A Convergent Reformulation of Adam with Accelerated Rate
Yaxin Yu, Long Chen, Zeyi Xu
Adam has achieved strong empirical success, but its theory remains incomplete even in the deterministic full-batch setting, largely because adaptive preconditioning and momentum ar…
Adam-SHANG: A Convergent Adam-Type Method for Stochastic Smooth Convex Optimization
Yaxin Yu, Long Chen, Minfu Feng
We propose Adam-SHANG, a Lyapunov-guided Adam-type method that couples momentum, adaptive preconditioning, and a curvature-aware correction through a more stable lagged-preconditio…
RTP-LLM: High-Performance Alibaba LLM Inference Engine
Boyu Tan, Jiarui Guo, Zongwei Lv +26
Large Language Models (LLMs) have revolutionized AI applications, but deploying them at scale presents significant challenges. We present RTP-LLM, a high-performance inference engi…
Efficient Reflectance Capture with a Deep Gated Mixture-of-Experts
Xiaohe Ma, Yaxin Yu, Hongzhi Wu +1
We present a novel framework to efficiently acquire near-planar anisotropic reflectance in a pixel-independent fashion, using a deep gated mixtureof-experts. While existing work em…
SHANG++: Robust Stochastic Acceleration under Multiplicative Noise
Yaxin Yu, Long Chen, Minfu Feng
Under the multiplicative noise scaling (MNS) condition, original Nesterov acceleration is provably sensitive to noise and may diverge when gradient noise overwhelms the signal. In…